The overall goal of this project is to investigate the use of MR spectroscopic and functional imaging methods to predict and assess chemotherapeutic response in brain tumors. Specifically, this work proposes to ascertain if new MR neuro-imaging techniques can discriminate between response and non-response to PCV chemotherapy in patients with anaplastic oligodendroglioma (AO) and anaplastic oligoastrocytoma (AOA). AO and AOA are unique in their chemosensitivity with approximately three-quarters of these tumors responding dramatically to the PCV regimen. Our preliminary studies show three major findings: 1) specific loss of chromosome 1p and 19q is strongly correlated with chemotherapeutic response in AO; 2) different tumor types give rise to quantifiable differences in the MRS metabolic pattern; 3) ex vivo MRS results accurately reflect in vivo MRS metabolic patterns. The discovery of the unique genetic correlates of chemosensitivity in AO and AOA presents a singular opportunity to investigate the relationship between genetics, MR detected metabolic phenotypes and chemotherapeutic response. We therefore hypothesize that metabolic and physiological MR imaging techniques can identify non-responsive tumors at an earlier time point than conventional imaging modalities. The following specific aims are designed to test this hypothesis: 1) To correlate in vivo and ex vivo MRS patterns and molecular genetics as predictors of response in untreated AO and AOA. 2) To characterize the in vivo spectroscopic patterns, rCBV maps and diffusion properties of AO and AOA prior to treatment and at specific time points during the initial cycle of chemotherapy. 3) To determine the relationship between MR metabolic, functional imaging and clinical response in patients with AO and AOA. The successful completion of this study will provide an assessment of the ability MR spectroscopic and functional imaging techniques to predict chemosensitivity prior to treatment and to monitor response throughout the therapy.